HAL Id: hal-02147845
https://hal.sorbonne-universite.fr/hal-02147845
Submitted on 5 Jun 2019
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Fate of 13 C labelled root and shoot residues in soil and anecic earthworm casts: A mesocosm experiment
A. Vidal, K. Quenea, M. Alexis, T.T. Nguyen Tu, J. Mathieu, V. Vaury, S.
Derenne
To cite this version:
A. Vidal, K. Quenea, M. Alexis, T.T. Nguyen Tu, J. Mathieu, et al.. Fate of 13 C labelled root and shoot residues in soil and anecic earthworm casts: A mesocosm experiment. Geoderma, Elsevier, 2017, 285, pp.9-18. �10.1016/j.geoderma.2016.09.016�. �hal-02147845�
1 Fate of 13C labelled root and shoot residues in soil and anecic earthworm casts: a 1
mesocosm experiment 2
A.Vidal a, K. Quenea a,*, M. Alexis a, T. T. Nguyen Tu a, J. Mathieu b, V. Vaury b, S. Derenne a 3
aUMR Milieux environnementaux, transferts et interactions dans les hydrosystèmes et les 4
sols (METIS), UMR 7619, UPMC, CNRS, EPHE, 4 place Jussieu, F-75252 Paris, France 5
bUMR IEES-Paris, UMR 7618, UPMC, UPEC, CNRS, INRA, IRD, AgroParisTech, 4 place 6
Jussieu, F-75252 Paris, France 7
*Corresponding author:
8
+33 1 44 27 42 21 9
Email address: [email protected] 10
Address: UPMC, Tour 56-66, 4 place Jussieu, 75252 Paris, France 11
12
2 Abstract
13
Earthworms are known to have a major impact on organic matter dynamics in soils. The pre- 14
cise dynamics of carbon incorporation and/or decomposition in soil under the influence of 15
earthworms still need to be investigated. In a mesocosm experiment, the fate of Ryegrass 16
root and shoot litter was monitored in the soil, in the presence and absence of anecic earth- 17
worms Lumbricus terrestris L. Residues were13C labelled and deposited onto the soil sur- 18
face. Incorporation of 13C in surface casts and in the 0-20 and 40-60 cm soil layers was 19
monitored 1, 2, 4, 8, 24 and 54 weeks after adding labelled litter. Organic carbon content and 20
δ13C values were obtained for all samples, allowing the determination of the percentage of 21
carbon derived from labelled litter (Clab). Roots and shoots were incorporated in the 0-20 cm 22
soil layer during the year of experiment, Clab reaching 11.4 % of the soil organic carbon after 23
54 weeks. On the contrary, no significant contribution from labelled residues was observed in 24
the 40-60 cm layer. Roots decomposed at a slower rate compared to shoots. Litter incorpora- 25
tion was observed in casts from the very first weeks of experiment (Clab from 34.8 to 51.4 % 26
after 2 weeks). In the soil, a significant effect of earthworms on the Clab was detected after 24 27
weeks. Earthworms accelerated root and shoot decomposition in the soil. They also en- 28
hanced, in the presence of shoot residues, the decomposition of the organic matter originally 29
present in the soil. However, after one year, earthworms smoothed the difference between 30
residue types in casts and to a lesser extent in soil, revealing their capacity to enhance the 31
decomposition of both roots and shoots.
32
Keywords 33
Organic matter dynamics 34
Carbon 35
Isotope labelling 36
Anecic earthworms 37
3 Casts
38
1. Introduction 39
Soil organic matter (SOM) is (1) a source of nutrients for soil organisms, (2) a key 40
determinant of soil fertility, structuration and evolution and (3) a potential sink of carbon (C) in 41
a context of CO2 concentration increase in the atmosphere (Jobbágy and Jackson, 2000).
42
SOM has been widely studied but the process of its incorporation and dynamics in soils 43
remains unclear due to its complexity (Stockmann et al., 2013). The main source of SOM is 44
plant residues, which encompass plant above-ground parts (dead leaves and shoots) 45
deposited on the soil surface and below-ground parts (dead roots and rhizodeposits) within 46
the soil (Kögel-Knabner, 2002). Residues can either be mineralized, releasing CO2 to the 47
atmosphere, or incorporated into soil in the form of organic compounds. After incorporation, 48
these compounds can be mineralized, transferred in deeper layers and/or stored in soil.
49
Residue decomposition and incorporation into soil depend on numerous abiotic and biotic 50
factors. Abiotic factors mainly correspond to soil climate (temperature and moisture) and 51
physico-chemical characteristics (texture and clay mineralogy). These factors have a critical 52
influence on organo-mineral interactions, impacting organic matter decomposition and 53
accumulation in soil. Biotic factors encompass residue quality, activity of soil fauna and 54
microorganisms (Cortez, 1998; Oades, 1988). Soil microorganisms, which comprise, among 55
others, bacteria and fungi, mediate the processes of organic compound mineralization and 56
transformation in soil. Thus, they have a direct implication on residue decomposition 57
(Kuzyakov and Blagodatskaya, 2015). Under relatively constant climatic and edaphic 58
conditions, abiotic factors often influence litter decomposition at large time and space scales, 59
while biotic factors act at shorter time and space scales (Lavelle et al., 1993; Swift et al., 60
1979), 61
Residue quality (i.e. physico-chemical features including chemical composition, anatomical 62
characteristics, etc.) influences their decomposition by macro and microorganisms (Lavelle et 63
4 al., 1993). While above-ground biomass has long been considered as the main source for 64
SOM, roots are currently recognized as essential contributors to stable organic C. The 65
relative contribution from root and shoot litter to SOM has been the subject of many studies 66
during the last decades (Balesdent and Balabane, 1996; Comeau et al., 2013; Gale and 67
Cambardella, 2000; Lu et al., 2003; Mambelli et al., 2011; Puget and Drinkwater, 2001;
68
Rasse et al., 2005) and it is now recognized that roots decompose at a slower rate compared 69
to shoots. However, the reasons explaining the slower decomposition of roots and its higher 70
contribution to soil carbon pool remain debated. Are they driven by the chemical composition, 71
physical protection and/or physico-chemical protection of roots in soils (Rasse et al., 2005)?
72
The contribution from these 3 factors on C dynamics is particularly dependent on residue 73
location in soil (Coppens et al., 2006a, b; Tahir et al., 2016). For example, Coppens et al.
74
(2006a) found that, after 9 weeks of experiment, new organic C storage in soil was increased 75
when rapeseed residues were deposited onto the soil surface compared to residues 76
incorporated in the 0-10 cm layer. In order to investigate the impact of the chemical 77
compositions of residues, the latter must be placed in the same conditions (Rasse et al., 78
2005). Thus, in the present study, both roots and shoots were deposited onto the soil 79
surface.
80
Among biotic factors, earthworms play a key role in soil structuration, residue decomposition 81
and C cycling (Blouin et al., 2013; Bossuyt et al., 2005). Earthworms feed on both mineral 82
(soil) and organic matter (litter, humus and microorganism), in different proportion depending 83
on their ecological category. For example, anecic earthworms, which represent the dominant 84
earthworm biomass in temperate regions, feed on litter deposited on the soil surface and 85
transfer it along vertical burrows which can reach 1-2 meters (Lee et al., 1985). During 86
ingestion, residues are fragmented and the preexisting soil microstructures destroyed.
87
Mineral and organic elements are mixed, complexed with mucus and released in the form of 88
organo-mineral aggregates called casts (Lee, 1985; Six et al., 2004). This process creates 89
nuclei for the formation of organo-mineral aggregates in soil.Thus, earthworms promote the 90
5 formation of biogenic macroaggregates (Bossuyt et al., 2005), which are generally more 91
stable than non-biogenic aggregates (Six et al., 2004; Zangerlé et al., 2011). Earthworms are 92
highly mobile in soil and induce heterogeneous resource distribution (Decaëns et al., 2010).
93
As the ingested residues contain higher concentration of organic C than soil, they modify the 94
incorporation and the stock of C in aggregates (Arai et al., 2013; Fonte et al., 2012, 2007;
95
Hong et al., 2011) and along the soil profile (Jégou et al., 2000). The net impact of 96
earthworms on soil C strongly varies depending on the studied spatial and time scale. During 97
gut transit and in fresh casts, mineralization of organic C is accelerated. Indeed, in 98
earthworm gut, the physical protections of the organic matter originally present in the soil are 99
broken and plant residues are fragmented (Lavelle and Martin, 1992). Moreover, the 100
presence of water and intestinal mucus creates conditions favorable to mineralization by the 101
microorganisms present in the earthworm gut and fresh casts (Brown et al., 2000; Drake and 102
Horn, 2007; Martin et al., 1987; Vidal et al., 2016a). At month or year scale, drying and 103
ageing casts, lead to C stabilization (Brown et al., 2000; Lavelle and Martin, 1992; Martin, 104
1991). Residue decomposition rate by earthworms also depends on residue palatability 105
(Cortez et al., 1989). For example, it has generally been observed that earthworms fed 106
mainly on shoots rather than roots (Bouché and Kretzschmar, 1974; Curry and Schmidt, 107
2007). However, little is known about the incorporation and decomposition of root residues 108
by earthworms in soil (Curry and Schmidt, 2007; Zangerlé et al., 2011).
109
Artificial isotopic labelling of residues allows precise monitoring of the residues introduced (at 110
a given date and place) and distinguishing the added organic matter from that initially present 111
in the soil (Brüggemann et al., 2011; Comeau et al., 2013; Fahey et al., 2013; Klumpp et al., 112
2007; Thompson, 1996). Studies using artificial labelling have focused on (1) the impact of 113
residue quality (roots vs. shoots or plant species) on organic C incorporation and 114
decomposition into soil (Blair et al., 2005; Comeau et al., 2013; Mambelli et al., 2011; Rubino 115
et al., 2010; Williams et al., 2006) or (2) the influence of earthworms on shoot decomposition 116
and incorporation into soil and the consequences on soil C storage (Fahey et al., 2013;
117
6 Fonte et al., 2007; Stromberger et al., 2012). The above mentioned studies have proven that, 118
taken separately, these two biotic factors (earthworm and residue quality) have a crucial 119
impact on soil C cycling. However, is the diverging fate of root and shoot residues in soil 120
modified in the presence of earthworms? In this study, a one year mesocosm experiment 121
was set up to monitor the fate of root and shoot residues in the presence and absence of 122
earthworms. Residues were artificially 13C labelled and deposited onto the soil surface.
123
7 2. Materials and Methods
124
2.1. Soil characteristics and mesocosm experiment 125
The surface layer of a soil was collected on a permanent grassland, largely dominated by 126
Ryegrass, in the North of France (Aux-Marais, Oise). The mean annual precipitation and 127
temperature in the region were 650 mm and 11°C, repectively. The soil collected is a loamy- 128
sand soil with characteristics presented in Table 1. The experiment was performed in six 129
PVC containers (length - 80 cm, diameter - 40 cm) filled with ca. 75 L of soil. It took place in 130
a greenhouse, under controlled conditions (air conditioning), with natural light and the soil 131
was kept at approximately 13°C and 23 % of humidity. The soil was homogenized and sieved 132
at 4 mm in order to obtain homogenous columns of soil. Before starting the experiment, the 133
soil containers were pre-incubated during 6 months in the same conditions as during the 134
experiment. This allowed passing the bacterial flush induced by drying-rewetting of soil (Van 135
Gestel et al., 1993; Franzluebbers et al., 2000) and/or soil physical disturbance during 136
sieving (Datta et al., 2014). The soil macrofauna was first removed manually during sieving 137
and remaining earthworms were extracted from the soil using an electrical device during pre- 138
incubation, as described in Weyers et al. (2008). Each container was weekly vaporized with 139
water (1L/week) and seedlings or mosses in mesocosms were eliminated.
140
The six treatments were identified as (1) Root-E for the mesocosm with root residues and 141
earthworms (2) Root-NE for the mesocosm with root residues and without earthworms (3) 142
Shoot-E, for the mesocosm with shoot and earthworms (4) Shoot-NE for the mesocosm with 143
shoot and without earthworms (5) Control-E, for the mesocosm without plant residues and 144
with earthworms and (6) Control-NE for the mesocosm without plant residues and without 145
earthworms (Fig.1).
146
2.2. Earthworms and root-shoot 147
The common anecic earthworm Lumbricus terrestris L. was used. Earthworms were provided 148
by the SARL Lombri’carraz (France) and were received at adult stage, with an average 149
8 weight of 4.5 g. They were pre-conditioned during one week in the soil used for the
150
experiment. At the beginning of the experiment, 6 earthworms per container were deposited 151
on the top of three of the mesocosms. Mean living mass was 26.8 ± 0.5 g per container 152
(around 48 ind./m2 and 224 g fresh wt/m2). This low density was chosen in accordance to 153
Fründ et al. (2010) review, showing that the typical earthworm density (all species 154
concerned) in temperate pastures was 300-1000 ind./m2 (50-100 g fresh wt/m2). Thus, as 155
L.terrestris is a territorial earthworm, these choices favored their survival in conditions closest 156
to reality. By the end of the experiment, the soil volume extracted during the year of sampling 157
represents approximatively 6 % of the initial soil volume. Thus, it is assumed that earthworms 158
were not limited in their movement along the experiment.
159
Plants of Italian Ryegrass (Lolium multiflorum Lam.) were artificially labelled with 13C at the 160
Atomic Energy and Alternative Energies Commission (CEA) in Cadarache (France). In order 161
to obtain homogeneous material, plants were grown under a controlled and constant 13CO2
162
enriched atmosphere. Immediately after 13C labelling, fresh roots and shoots were separated, 163
dried and chopped. They were chopped during 40 seconds with a laboratory blender (Waring 164
Commercial) in order to obtain residues of few millimeters. The mean δ13C values were 1324 165
‰ (± 43) and 1632 ‰ (± 16) for the roots and shoots, respectively (Table 2). Few hours after 166
earthworms had penetrated the soil, roots and shoots were similarly applied on the soil 167
surface. Thus, the impact of the chemical composition of roots vs shoots was tested, 168
excluding any influence of their location in the soil. Two mesocosms each contained 250 g of 169
shoots, two additional contained 250 g of roots and the latest two were used as control.
170
Curry and Schmidt (2007) reviewed that L.terrestris could ingest 17 mg of dry matter/g fresh 171
wt/day. However, the quantity of litter ingested can increase when the quality of litter 172
decreases. Thus, 250 g of litter (i.e. around 25 mg of dry matter/g fresh wt/day) was chosen 173
as a good compromise to favor earthworm survival.
174
2.3. Sampling and analyses 175
9 Samples were collected at seven time steps: before and 1, 2, 4, 8, 24 and 54 weeks after the 176
addition of roots or shoots and earthworms. At each time step and for each mesocosm, three 177
replicates of soil samples were collected at 0-20 and 40-60 cm with an aluminum auger 178
(height: 100 cm, diameter: 2.5 cm), after putting aside plant residues. Few grams of 3-4 179
earthworm cast fragments were collected on the soil surface in the mesocosms containing 180
earthworms in order to obtain one composite sample for each time step. During the first 6 181
months of experiment, most casts were fresh when collected. Casts collected after 54 weeks 182
were starting ageing and drying. The casts collected in the mesocosms Root-E, Shoot-E and 183
Control-E will be identified as Cast-Root, Cast-Shoot and Cast-Control, respectively. All 184
samples were freeze-dried and ground. After 54 weeks, roots and shoots remaining at the 185
soil surface (i.e. for Root-NE and Shoot-NE) were collected and weighted.
186
2.4. Organic carbon and δ13C analyses 187
Organic C (Corg) and δ13C were measured on all soil replicates with Elemental Analysis – 188
Isotope Ratio Mass Spectrometry (EA-IRMS) Vario pyro cube – Micromass Isoprime. The 189
standard gas was calibrated in relation to the international standard (PeeDee Belemnite- 190
PDB). For the casts, the analysis was repeated three times. All samples were decarbonated 191
before the analysis, using the method developed by Harris et al. (2001). Briefly, subsamples 192
of soil (~40 mg) and casts (~15 mg) were placed in open silver capsules (8*5 mm). Samples 193
were placed in a microtiter plate and 15 µL of ultra-pure water were added. They were placed 194
in a vacuum desiccator containing a 100 mL beaker of HCl (12 M). Samples were exposed to 195
HCl vapor during 6 hours and then placed under vacuum, without HCl, to remove HCl 196
potentially trapped in the sample. Samples were then transferred to an oven at 40°C for 15 197
hours.
198
Isotopic composition of samples is expressed in δ13C relative to the PDB standard (Equation 199
200 1):
δ13C (‰) = [(Rsample – Rstandard) / Rstandard] * 1000 (1) 201
10 Where R is the 13C/12C ratio of the sample or the standard.
202
The proportion of C derived from the labelled litter in the soil or casts (Clab), was expressed in 203
percentage by applying equation 2:
204
Clab (%) = [(δs – δc) / (δl – δc)] * 100 (2) 205
Where δs is the δ13C value of soil samples or of casts with labelled litter, δc is the δ13C value 206
of the control soil sample or control casts without litter, δl is the δ13C value of the labelled 207
litter.
208
In order to overcome the initial variability of the soil organic C measured in the different mes- 209
ocosms, all the organic C contents were normalized using equation 3:
210
RCorg x = Cx / C0 (3) 211
Where RCorg x represents the trend in organic C content ratio of C after x weeks, Cx is the 212
organic C content after x weeks of experiment and C0 is the soil organic C content before 213
adding earthworms and residues, for the corresponding mesocosm. Thus, values lower or 214
higher than one express a decrease or increase in organic C content in comparison to the 215
beginning of the experiment, respectively.
216
A similar formula was applied for the casts, where Cx is the organic C content in cast and C0
217
is the organic C content in the soil before adding earthworms and residues, for the 218
corresponding mesocosm. Thus, values higher than one indicate that the organic C content 219
is higher in the casts compared to the initial soil.
220
2.5. Statistical analysis 221
Statistical analyses were realized using the R statistical software. As data were not normally 222
distributed (Shapiro-Wilk test), samples from different mesocosms were compared using the 223
non-parametric Kruskal-Wallis test followed by a pairwise comparison of groups from the 224
package ‘RcmdrPlugin.coin’. For all analyses, statistical significance was set as α = 0.05.
225
11 3. Results
226
Observation of mesocosms during the year of experiment gave first indications on the incor- 227
poration of the labelled residues. After 24 weeks of experiment, no more shoot residues were 228
observed in the Shoot-E mesocosm whereas, in the Root-E mesocosm, root residues were 229
still observed on the soil surface, occupying less than 10 % of the surface. After one year of 230
experiment, no more residues were observed at the soil surface in the mesocosm containing 231
earthworms. On the contrary, in the absence of earthworms, after one year, some remaining 232
residues were still visible at the soil surface, for either root or shoot residues. They were 233
roughly separated from the soil and weighted. They accounted for ca. 30 % of the initial 234
weight. After one week, casts were observed in Shoot-E. However, no casts were identified 235
until two weeks in the mesocosm with root residues. In spite of regular removal of seedling 236
and mosses during the experiment, some mosses developed in the Control-NE mesocosm, 237
from 24 weeks until the end of the experiment.
238
3.1. Carbon derived from root and shoot residues 239
3.1.1. Soil samples 240
The C derived from labelled litter (Clab) increased continuously in the 0-20 cm soil layer along 241
the year of experiment, reaching between 0.96 and 11.4 % after 54 weeks, for Shoot-E and 242
Root-NE, respectively (Fig.2A). From the beginning of the experiment to the 8th week, there 243
were no significant differences between mesocosms, with a mean Clab of 0.40 %. After 24 244
weeks, Clab was significantly lower for Shoot-E (0.85 %) than Root-E (3.00 %). After 54 245
weeks, Clab was significantly higher in Root-NE (11.4 %) than in the other mesocosms. Thus, 246
two general trends could be observed after 24 and 54 weeks: (1) Clab in Root-NE and Shoot- 247
NE tended to be higher than Root-E and Shoot-E, respectively and (2) Clab in Root-E and NE 248
tended to be higher than Shoot-E and NE, respectively. Contrary to other mesocosms, Clab in 249
Shoot-E barely varied from 8th week until the end of experiment, with values of 1.13, 0.85 and 250
0.96 % after 8, 24 and 54 weeks, respectively. There was no statistically significant differ- 251
12 ence of Clab between mesocosms in the 40-60 cm soil layer during the year of experiment 252
(Fig.2B). However, at 54 weeks Clab reached 0.2 % in Root-E and Shoot-E whereas it re- 253
mained lower than 0.05 % without earthworms.
254
3.1.2. Earthworm casts 255
For all the time steps, Clab measured in casts was higher than Clab in soil, with a mean Clab of 256
37 % with respect to 1.3 % for soils, all time steps and mesocosms included (Fig.3). Clab in 257
Cast-Shoot reached 34.9 % in average during the first month of experiment. Clab reached a 258
maximum after 8 weeks for shoots (58.1 %) and after 8 and 24 weeks for roots (51.8 %). Clab
259
then decreased during the last six months of experiment, reaching around 12 % for both 260
Cast-Root and Cast-Shoot after 54 weeks.
261
3.2. Organic carbon content in soil and cast samples 262
3.2.1. Soil samples 263
Due to the very low litter C incorporation for the 40-60 cm soil layer, the following of this sec- 264
tion is focused on the 0-20 cm layer. In mesocosms containing shoots, with or without earth- 265
worms, organic C content tended to be lower than the organic C content before the experi- 266
ment (RCorgx ≤ 1), for all time steps (Fig.4 and Table S1). On the contrary, in mesocosms 267
containing roots, with or without earthworms, organic C content tended to be higher than at 268
the beginning of the experiment (RCorg x≥ 1). RCorg reached 1.2 after 54 weeks in Root-NE.
269
RCorg was significantly different for Root-E and Shoot-E after 24 weeks (1.1 vs 0.85).This 270
observation was also made between Root-NE and Shoot-NE but to a lesser extent. For ex- 271
ample, after 24 weeks, RCorg was 1.1 and 0.89 for Root-NE and Shoot-NE, respectively 272
(Fig.4).
273
3.2.2. Earthworm casts 274
The organic Cmeasured in casts was always higher than the soil organic C measured before 275
the experiment (RCorg >>1) (Fig.5). For Cast-Control, the RCorg wasstable for all the time 276
13 steps, with a mean value of 1.3. Both Cast-Root and Cast-Shoot, RCorg remained stable at 277
ca. 1.8, during the first month of experiment before increasing to their maximum at 8 weeks 278
and subsequently decreasing until the end of the experiment. When Cast-Root and Cast- 279
Shoot are compared, their RCorg are similar during the first month and it becomes higher for 280
Cast-Root after 8 (3.6 vs 2.7) and 24 (2.8 vs 2.3) weeks of experiments before reaching 281
again similar values after 54 weeks (1.5 and 1.6).
282
4. Discussion 283
The proportion of C derived from the labelled litter into the 0-20 cm soil layer after one year 284
of experiment (Clab between 0.96 and 11.4 % depending on the mesocosm, (Fig.2A)) is 285
consistent with previous studies (Kammer and Hagedorn, 2011; Rubino et al., 2010).
286
Residues placed on the soil surface have either been (1) totally mineralized, (2) left intact at 287
the soil surface and/or (3) transferred to the soil either in form of soluble compounds or as 288
undecomposed fragmented pieces (Cortez and Bouché, 1998; Rubino et al., 2010). Thus, 289
the 13C signal measured in soil could either derived from the residues themselves (soluble or 290
small fragments) or from microorganisms that have degraded the residues. No significant 291
differences are observed between the mesocosms until the 24th week of experiment. The 292
absence of significant earthworm impact during the first 8 weeks can be explained by the 293
experimental conditions. Indeed, the low density of earthworms and the localized sampling to 294
preserve soil structure in mesocosms, accounted for the great variability of data and the late 295
significant impact of earthworms. Differences induced by root or shoot residues were not 296
detected until 6 months. This could be explained by the experimental conditions. Indeed, as 297
high volumes of soil were used in the present study, an important quantity of labelled C was 298
needed to reveal a significant difference between samples.
299
4.1. Root vs. shoot effect 300
For the last two time periods, the Clab of mesocosms containing roots was higher than those 301
with shoots, although the difference was not systematically statistically significant (Fig 2.A).
302
14 The higher contribution to SOM of root compared to shoot residues was also supported by 303
the soil RCorg results after 54 weeks in Root-NE and Shoot-NE (1.20 and 1.05, respectively).
304
These results are coherent with William et al. (2006) who found, after having incubated roots 305
and shoots of ryegrass in soil for 10 months, a C contribution of 5.5 % and 1.5 %, 306
respectively. Two explanations can be put forward to explain the higher root-derived C input 307
in soil. First, roots could be more transferred from residues into soil (as dissolved OM or 308
small fragments), compared to shoots. However, this explanation is very unlikely, considering 309
the characteristics of roots compared to shoots, as detailed in the second explanation.
310
Second, root-derived C decomposed at a slower rate than shoot-derived C, as evidenced in 311
numerous studies (Beuch et al., 2000; Gale and Cambardella, 2000; Scheu and 312
Schauermann, 1994; Shi et al., 2013; Steffens et al., 2015). Indeed, the higher mineralization 313
of shoots compared to roots can be explained by (1) the higher water soluble C 314
concentration in shoots (Beuch et al., 2000; Bird and Torn, 2006; Shi et al., 2013), which is 315
rapidly transferred to soil and decomposed by microorganisms (Shi et al., 2013); (2) the 316
higher chemical recalcitrance of roots due to higher concentration of resistant molecules 317
such as lignin, tannins and suberins (Gleixner et al., 2001; Kögel-Knabner, 2002; Rasse et 318
al., 2005). The higher relative abundance of lignin-derived compounds in the roots used in 319
the present study compared to shoots, was confirmed by Vidal et al. (2016b) using molecular 320
characterization and (3) The lower C/N ratio of shoots vs. roots (15 vs. 31 for the litter used 321
in this study – Table 2), which is generally related to an increased rate of litter decomposition 322
(e.g. Nguyen Tu et al., 2011; Taylor et al., 1989; Witkamp, 1966).Thus, the present results 323
highlight the importance of the chemical composition on residue decomposition. It can be 324
suggested that the difference between root and shoot degradation is mainly driven by soluble 325
carbon content at short term (first weeks of experiments), whereas, the chemical 326
recalcitrance of roots probably becomes the dominant factor at longer terms, as suggested in 327
the literature (e.g.Trinsoutrot et al., 2000; Machinet et al., 2011).
328
4.2. Effect of earthworms on residue fate 329
15 4.2.1. Earthworms tended to accelerate residue incorporation and decomposition
330
No more residues were observed at the soil surface after the year of experiment in the 331
mesocosm containing earthworms compared to the one without earthworms. Thus, residues 332
had probably been incorporated in soil or casts and/or mineralized. Although absolute 333
quantification was not achieved, it can reasonably be suggested that earthworms accelerated 334
decomposition of residues in soil. Thus, part of the 13C was probably removed from the 335
system in the form of 13CO2 (microbial and/or earthworm respiration). Indeed, from 6 months 336
of experimentation, the Clab in soil tended to be lower in the presence of earthworms, for both 337
root and shoot residues (Fig.2A). This can be explained by the direct impact of earthworms 338
which assimilate and mineralize part of the labelled organic C during the ingestion of 339
residues, as suggested by Daniel (1991). Moreover, decomposition and mineralization are 340
governed by microorganisms acting directly on particulate residues, or on soluble 341
compounds in the soil matrix (Gaillard et al., 1999). In the presence of earthworms, this 342
phenomenon can be amplified as the ingestion of residues by earthworms generally favors 343
microbial activity (Amador and Görres, 2007; Brown, 1995; Frouz et al., 2011; Parle, 1963;
344
Vidal et al., 2016a).
345
4.2.2. Casts – a key role in residue decomposition 346
Cast and soil samples presented different trends of organic C incorporation and 347
decomposition (fig. 2.A vs. 3 and fig 4 vs. 5). As expected, the Clab and Corg are higher in 348
casts compared to soil. The organic C content of casts was 1.3 to 3.6 higher compared to the 349
soil. Similarly, Mariani et al. (2007) observed that the C content in anecic casts collected in 350
burrows of a Colombian pasture, was twice higher compared to bulk soil. This can be 351
explained by the feeding behavior of earthworms, especially anecic ones, which incorporate 352
fresh residues and/or organic-rich compounds in casts (Decaëns et al., 1999; Frouz et al., 353
2011; Hong et al., 2011; Zangerlé et al., 2011; Zhang et al., 2003). The maximum value for 354
Clab (58.1% in Cast-Shoot, after 8 weeks of incubation) is in agreement with a previous study 355
16 which reported that half of the C in L.terrestris casts was derived from labelled litter after 246 356
days of experiment, adding labelled litter every day (Jégou et al., 1998).
357
A chronology can be proposed, explaining residue incorporation by earthworms in casts 358
(Figure 6), based on Clab and RCorg x results (Fig. 3 and 5). After a strong increase within the 359
first week of experiment, both Clab and RCorg in casts remained stable during the four 360
following weeks. In this phase, earthworms have probably combined direct and indirect 361
incorporation of residues. During direct incorporation, residues are ingested and pass 362
through the earthworm gut (Shipitalo and Protz, 1989). Indirect incorporation, also called 363
‘pre-oral litter decomposition’, consists in a ‘ploughing-in’ of residues, i. e. casts are 364
deposited on residues which are not ingested, initiating microbial decomposition (Cortez and 365
Bouché, 1998). An increase in Clab and RCorg was observed between 4 and 8 weeks. After 366
few months, earthworms are able to re-ingest their casts in which residues have been 367
partially decomposed by microorganisms (Swift et al., 1979; Jiménez et al., 1998; Brown et 368
al., 2000). Thus, between 4 and 8 weeks of experiment, the ‘ploughing-in’ and ingestion of 369
undecomposed labelled residues still present at the soil surface were probably associated to 370
the re-ingestion of previously labelled casts (from 1 to 4 weeks), increasing the Clab. During 371
the last 6 months, Clab and RCorg highly decreased, no more litter was present at the surface 372
of the mesocosms containing earthworms and litter, and surface casts were starting ageing 373
and drying. Thus, all the residues have been either incorporated in the soil and casts, or 374
mineralized. The decrease in Clab and RCorg indicated that labelled residues incorporated in 375
casts sampled at 54 weeks have been partially mineralized. This decomposition was 376
enhanced by bacteria and fungi as observed on Cast-Shoot NanoSIMS images after 24 377
weeks of experiment (Vidal et al., 2016a).
378
4.2.3. A slight effect in the 40-60 cm soil layer 379
A slight increase in Clab was observed in the 40-60 cm soil layer during the 54 weeks in the 380
presence of earthworms, although not statistically significant (Fig.2B). This was observed in 381
17 other studies which found that below 2 or 5 cm, labelled litter incorporation was not
382
significant, at least for short time scales (Jégou et al., 1998; Kammer and Hagedorn, 2011;
383
Rubino et al., 2010), mainly due to the limited migration of soluble C in deeper soil layers 384
(Thompson, 1996). However, the burrows of L. terrestris are considered as permanent 385
structures that can reach deep soil layers (>1 m) (Jégou et al., 2000; Don et al., 2008). Thus, 386
earthworms may have contributed to the incorporation of labelled organic compounds into 387
the 40-60 cm soil layer directly, by covering their burrow walls with mucus and belowground 388
casts, both enriched in 13C (Jégou et al., 2000) or to a lesser extent, indirectly, by creating 389
preferential paths for labelled water soluble C circulation (Don et al., 2008). The Clab increase 390
was probably not significant because earthworm influence is localized on burrow walls and 391
their periphery, and casts (Don et al., 2008). Jegou et al. (2000) showed in an experiment 392
using 13C labelled litter that below the 2 cm soil layer, the C litter enrichment in the soil 393
surrounding burrows was barely detectable. Thus, the 13C signal was probably diluted in the 394
soil as earthworm burrows were not sampled separately.
395
4.3. Effect of earthworms on the fate of root vs. shoot 396
4.3.1. Earthworms minimized the diverging fate of root and shoot residues 397
In soil samples, the difference in Clab and RCorg between root and shoot mesocosms with 398
earthworms, after 54 weeks, was weaker than in the absence of earthworms (Fig.2A and 3), 399
underlying their ability to enhance root decomposition, as mentioned by Cortez and Bouche 400
(1992). Hopkins et al. (2005) have compared decomposition of natural tobacco roots and 401
modified tobacco roots with altered lignin structure and composition (i.e. less resistant to 402
decomposition). They found that earthworms reduced the difference between both residue 403
types for C mineralization. Earthworms can enhance the decomposition of recalcitrant 404
residues and smooth the difference between residues of various qualities, probably by 405
facilitating the access of plant residues to microorganisms. Indeed, a mutualistic relation is 406
maintained between bacteria and earthworms in their gut, which enabled the decomposition 407
of highly complex molecules (Barois and Lavelle, 1986; Trigo and Lavelle, 1993).
408
18 The difference between root and shoot residues was also weak in casts. Cast-Root and 409
Cast-Shoot both exhibited similar evolutionary trends for Clab and RCorg x (Figs. 3 and 5).
410
However, the observation of the mesocosms at the different time steps revealed some 411
differences. After one week of experiment, no cast could be observed on the soil surface in 412
the mesocosm containing roots and the litter appeared unaffected. Moreover, after 24 weeks, 413
while shoots had totally disappeared from the soil surface in Shoot-E, roots were still visible 414
in Root-E. Thus, in our experimental conditions, dead roots in early stage of decomposition 415
seemed less palatable for earthworms than shoots, confirming observations of Zangerlé et 416
al. (2011). Coq et al. (2007) also observed that less casts were produced with woody twigs of 417
soybean compared with small straws of rice, underlining the importance of residue 418
palatability for earthworm ingestion. This can be partly explained by the high C/N ratio of 419
roots in comparison to shoots as it is known that earthworms consume preferentially organic 420
compounds with low C/N (Curry and Schmidt, 2007). However, after 2 weeks, the Clab
421
increases rapidly in Cast-Root, underlying the capacity of earthworms to process roots 422
(Cortez and Bouche, 1992). After 54 weeks, no more litter was observable in both 423
mesocosms and the Clab and RCorg were not significantly different for casts containing roots 424
or shoots (Figs 3 and 5). Thus, even though earthworms prefer feeding on shoots, they are 425
able to incorporate roots in casts, where the decomposition is activated at a comparable rate 426
to that observed with shoots.
427
4.3.2. Fate of organic matter initially present in soil 428
In the presence of earthworms, during the first 6 months, the RCorg x is higher in Root-E than 429
in Shoot-E (Table S1), with a significant difference after 6 months of experiment (Fig. 4).
430
Moreover, the Corg content in Shoot-E is generally lower than that measured before the 431
experiment in the same mesocosm (RCorg x ≤ 1). This difference likely reflects a priming effect 432
in this mesocosm, leading to a high mineralization of the organic matter originally present in 433
the soil. This is also observed to a lesser extent with shoots without earthworms. The 434
enhanced SOM mineralization in the presence of fresh residues has been reported in many 435
19 studies (Kuzyakov, 2010). Indeed, microorganisms depend on the energy present in fresh 436
residues to decompose the recalcitrant organic matter present in the soil (Fontaine et al., 437
2011, 2007). For example, Lu et al. (2003) observed that the simple presence of fresh 438
residues, either root or shoot, favored the degradation of native soil organic matter after 8 439
weeks of experiment, probably due to the stimulation of the microbial activity. The soil fauna, 440
including earthworms, was also mentioned as inducing a priming effect in soil by Lavelle et 441
al. (1995). More recently, Coq et al. (2007) showed that total soil C had decreased after 5 442
months of incubation of rice or soybean shoot residues, in the presence of endogeic 443
earthworms. Fontaine et al. (2011) measured the effect of a 13C labelled cellulose input in a 444
soil during 161 days. They showed that, after one month, respiration of unlabelled CO2
445
reached a maximum, revealing a priming effect. A continuous and slow decrease in 446
unlabelled CO2 respiration was then observed until the end of the experiment. This decrease 447
in priming effect is in agreement with the present results showing that, after one year, RCorg
448
was around 1 for Shoot-E and Shoot-NE. The suggested priming effect observed for the 449
mesocosm containing shoots was not observed for the one containing roots. Indeed, during 450
the first 6 months of experiment, the Corg tended to be higher in mesocosm containing roots 451
than at the beginning of the experiment (RCorg ≥ 1) Thus, in the present study, the priming 452
effect induced by root residues seems to be lower compared to that of shoot residues during 453
the first six months of experiment. This priming might also be compensated by the higher 454
preservation of root-derived carbon coupled to reduced activation of microbial activity 455
compared to shoot residues (Shi et al., 2013). This process is smoothed after one year;
456
supporting comment from Rasse et al. (2005) that the preferential shoot priming effect on 457
SOM mineralization cannot explain the higher root-derived C in soil, as observed after one 458
year on Fig.2A. It has to be noted that the increase in RCorg observed for Control-NE after 24 459
and 54 weeks can be explained by the development of mosses in this mesocosm from 24 460
weeks until the end of the experiment, potentially increasing the Corg concentration in the 0- 461
20 cm soil layer.
462
20 5. Conclusion
463
This mesocosm experiment allowed monitoring the fate of root and shoot residues in soil and 464
earthworm casts. Residues deposited onto the soil surface were continuously incorporated in 465
the 0-20 cm soil layer during the year of experiment. In agreement with previous studies, root 466
residues incorporated in the 0-20 cm soil layer tended to decompose at a slower rate than 467
shoot residues, underlining the essential role of litter chemical composition in the organic C 468
fate. Earthworms tended to accelerate both root and shoot residue decomposition in soil. The 469
activity of earthworms also induced a slight, but not statistically significant, transfer of C 470
derived from residues in the deeper soil layer (40-60 cm), without any distinction between 471
roots and shoots. In earthworm casts, root and shoot residues presented similar 472
incorporation and decomposition of Corg trends. A chronology for cast formation and evolution 473
was proposed, based on 13C and organic C results. Three phases were identified: (1) the first 474
month: mainly incorporation of fresh residues in casts, (2) until 6 months: re-ingestion of 475
labelled casts and incorporation of fresh residues and finally (3) after 6 months: mainly 476
microbial decomposition of residues in casts. This study provides new elements on the role 477
of earthworms on incorporation and decomposition of organic C from roots and shoots into 478
soil. Earthworms minimized the diverging fate of root and shoot residues, after the year of 479
experiment, in both soil and cast samples. Thus, the presence of earthworms tended to 480
enhance the decomposition of structures recognized as more chemically resistant. Moreover, 481
during the first six months of experiment, the presence of earthworms and shoot residues 482
enhanced the decomposition of soil organic C initially present in the soil. After one year, 483
earthworms had accelerated the incorporation and decomposition of plant residues in the 484
soil. However, after this period, the 13C signal had not decreased in the soil and had only 485
started to decrease in casts. Thus, longer term experiments are needed to investigate the 486
stabilization of C-derived from root and shoot residues in the presence of earthworms.
487
Moreover, while these results bring new light on the fate of root and shoot litter in the 488
21 presence of earthworms, they further require field-based experiment, including other
489
earthworm ecological categories, to be generalized.
490
Acknowledgements 491
Patrick Dumont is acknowledged for the management of the UPMC greenhouse. This study 492
benefits from the financial support from the EC2CO program from the French National 493
Institute of Sciences of the Universe (CNRS / INSU). We thank anonymous reviewers for 494
their constructive comments.
495
496
22 Table 1 – Bulk features of the soil used for the experiment
497
Soil characteristics Values Clay (< 2 μm) 189 g.kg-1 Loam (2-50 μm) 248 g.kg-1 Sand (50-2000 μm) 563 g.kg-1 Total carbonates (CaCO3) 19.0 g.kg-1 CEC Metson 9.90 cmol.kg-1 Organic Carbon 12.1 g.kg-1
Nitrogen 1.30 g.kg-1
δ13C -28.1 ‰
498
499
23 Table 2 – Initial litter characterization (n=13). Numbers in parentheses indicate the standard 500
deviation for the 13 samples.
501
δ13C (‰) C (g.kg-1) N (g.kg-1) C:N Root 1324 (42) 402 (40) 13.1 (1.2) 30.7 (3.1) Shoot 1632 (16) 408 (6.3) 27.7 (2.3) 14.8 (1.1) 502
503
24 Fig. 1. Experimental design. E: with earthworms, NE: without earthworms
504
Fig. 2. Proportion of carbon derived from labelled litter (Clab) measured in the soil samples at 505
the A. 0-20 cm and B. 40-60 cm layer, for the seven time steps, in mesocosms with labelled 506
residues (n=3). The scale is not the same for fig.2A and B. A significant difference between 507
mesocosms was revealed after 24 weeks, thus the letters above histogram bars represent 508
statistical results of Kruskal-Wallis test; different letters indicating significant difference be- 509
tween mesocosms, for a given time step. E: with earthworms, NE: without earthworms 510
Fig. 3. Proportion of carbon derived from labelled litter (Clab) in casts sampled in mesocosms 511
containing root (Cast-Root) or shoot residues (Cast-Shoot) (n=3). Letters above histogram 512
bars represent statistical results of Kruskal-Wallis test, different letters indicating significant 513
difference between mesocosms, for a given time step. No casts were identified until two 514
weeks in the mesocosm with root residues, explaining the absence of data for Cast-Root 515
after one week.
516
Fig. 4. Organic carbon content relative to the soil organic carbon content at the beginning of 517
the experiment (RCorg) (n=3), for the 0-20 cm soil layer and after 24 and 54 weeks. The let- 518
ters above histogram bars represent statistical results of Kruskal-Wallis test; different letters 519
indicating significant difference between mesocosms, for a given time step.
520
Fig. 5. Organic carbon content relative to the soil organic carbon at the beginning of the ex- 521
periment (RCorg) measured in casts sampled in the mesocosms containing no litter (Cast- 522
Control), root (Cast-Root) or shoot residues (Cast-Shoot) (n=3). The letters above histogram 523
bars represent statistical results of Kruskal-Wallis test; different letters indicating significant 524
differences between mesocosms, for a given time step. No casts were identified until two 525
weeks in the mesocosm with root residues, explaining the absence of data for Cast-Root 526
after one week.
527
25 Fig. 6. Simplified representation of the proposed chronology explaining residue incorporation 528
by earthworms within cast, in the present study. It has to be noted that most processes may 529
be combined for a same time step. Only the suggested dominant processes are represented.
530
531
References 532
533
Amador, J.A., Görres, J.H., 2007. Microbiological characterization of the structures built by 534
earthworms and ants in an agricultural field. Soil Biol. Biochem. 39, 2070–2077.
535
Arai, M., Tayasu, I., Komatsuzaki, M., Uchida, M., Shibata, Y., Kaneko, N., 2013. Changes in soil 536
aggregate carbon dynamics under no-tillage with respect to earthworm biomass revealed by 537
radiocarbon analysis. Soil Tillage Res. 126, 42–49.
538
Balesdent, J., Balabane, M., 1996. Major contribution of roots to soil carbon storage inferred from 539
maize cultivated soils. Soil Biol. Biochem. 28, 1261–1263.
540
Barois, I., Lavelle, P., 1986. Changes in respiration rate and some physicochemical properties of a 541
tropical soil during transit through Pontoscolex corethrurus (glossoscolecidae, oligochaeta).
542
Soil Biol. Biochem. 18, 539–541.
543
Beuch, S., Boelcke, B., Belau, L., 2000. Effect of the Organic Residues of Miscanthus × giganteus on 544
the Soil Organic Matter Level of Arable Soils. J. Agron. Crop Sci. 184, 111–120.
545
Bird, J., Torn, M., 2006. Fine Roots vs. Needles: A Comparison of 13C and 15N Dynamics in a 546
Ponderosa Pine Forest Soil. Biogeochemistry 79, 361–382.
547
Blair, N., Faulkner, R., Till, A., Sanchez, P., 2005. Decomposition of 13C and 15N labelled plant residue 548
materials in two different soil types and its impact on soil carbon, nitrogen, aggregate 549
stability, and aggregate formation. Aust J Soil Res 43, 873–886.
550
Blouin, M., Hodson, M.E., Delgado, E.A., Baker, G., Brussaard, L., Butt, K.R., Dai, J., Dendooven, L.., 551
Peres, G., Tondoh, J.. E., Cluzeau, D., Brun, J.-J., 2013. A review of earthworm impact on soil 552
function and ecosystem services. Eur. J. Soil Sci. 64, 161–182. doi:10.1111/ejss.12025 553
Bossuyt, H., Six, J., Hendrix, P.F., 2005. Protection of soil carbon by microaggregates within 554
earthworm casts. Soil Biol. Biochem. 37, 251–258.
555
Bouché, M., Kretzschmar, A., 1974. Fonction des lombriciens; II: Recherches méthodologiques pour 556
l’analyse du sol ingéré (étude du peuplement de la station RCP-165/PBI). Rev Ecol Biol Sol 11, 557
127–139.
558
Brown, G., 1995. How do earthworms affect microfloral and faunal community diversity? Plant Soil 559
170, 209–231.
560
Brown, G.G., Barois, I., Lavelle, P., 2000. Regulation of soil organic matter dynamics and microbial 561
activityin the drilosphere and the role of interactionswith other edaphic functional domains.
562
Eur. J. Soil Biol. 36, 177–198.
563
Brüggemann, N., Gessler, A., Kayler, Z., Keel, S.G., Badeck, F., Barthel, M., Boeckx, P., Buchmann, N., 564
Brugnoli, E., Esperschütz, J., Gavrichkova, O., Ghashghaie, J., Gomez-Casanovas, N., Keitel, C., 565
Knohl, A., Kuptz, D., Palacio, S., Salmon, Y., Uchida, Y., Bahn, M., 2011. Carbon allocation and 566
carbon isotope fluxes in the plant-soil-atmosphere continuum: a review. Biogeosciences 8, 567
3457–3489.
568
Comeau, L.-P., Lemke, R.L., Knight, J.D., Bedard-Haughn, A., 2013. Carbon input from 13C-labeled 569
crops in four soil organic matter fractions. Biol. Fertil. Soils 49, 1179–1188.
570
Coppens, F., Garnier, P., De Gryze, S., Merckx, R., Recous, S., 2006a. Soil moisture, carbon and 571
nitrogen dynamics following incorporation and surface application of labelled crop residues 572
in soil columns. Eur. J. Soil Sci. 57, 894–905.
573
26 Coppens, F., Merckx, R., Recous, S., 2006b. Impact of crop residue location on carbon and nitrogen 574
distribution in soil and in water-stable aggregates. Eur. J. Soil Sci. 57, 570–582.
575
Coq, S., Barthès, B.G., Oliver, R., Rabary, B., Blanchart, E., 2007. Earthworm activity affects soil 576
aggregation and organic matter dynamics according to the quality and localization of crop 577
residues—An experimental study (Madagascar). Soil Biol. Biochem. 39, 2119–2128.
578
Cortez, J., 1998. Field decomposition of leaf litters: relationships between decomposition rates and 579
soil moisture, soil temperature and earthworm activity. Soil Biol. Biochem. 30, 783–793.
580
Cortez, J., Bouché, M.B., 1998. Field decomposition of leaf litters: earthworm–microorganism 581
interactions —the ploughing-in effect. Soil Biol. Biochem. 30, 795–804.
582
Cortez, J., Bouche, M.B., 1992. Do earthworms eat living roots? Soil Biol. Biochem. 24, 913–915.
583
Cortez, J., Hameed, R., Bouché, M.B., 1989. C and N transfer in soil with or without earthworms fed 584
with 14C- and 15N-labelled wheat straw. Soil Biol. Biochem. 21, 491–497.
585
Curry, J.P., Schmidt, O., 2007. The feeding ecology of earthworms – A review. Pedobiologia 50, 463–
586
477.
587
Daniel, O., 1991. Leaf-litter consumption and assimilation by juveniles of Lumbricus terrestris L.
588
(Oligochaeta, Lumbricidae) under different environmental conditions. Biol. Fertil. Soils 12, 589
202–208.
590
Datta, R., Vranová, V., Pavelka, M., Rejšek, K., Formánek, P., 2014. Effect of soil sieving on respiration 591
induced by low-molecular-weight substrates. Int. Agrophysics 28, 119–124.
592
Decaëns, T., 2010. Macroecological patterns in soil communities. Glob. Ecol. Biogeogr. 19, 287–302.
593
Decaëns, T., Rangel, A.F., Asakawa, N., Thomas, R.J., 1999. Carbon and nitrogen dynamics in ageing 594
earthworm casts in grasslands of the eastern plains of Colombia. Biol. Fertil. Soils 30, 20–28.
595
Don, A., Steinberg, B., Schöning, I., Pritsch, K., Joschko, M., Gleixner, G., Schulze, E.-D., 2008. Organic 596
carbon sequestration in earthworm burrows. Soil Biol. Biochem. 40, 1803–1812.
597
Drake, H.L., Horn, M.A., 2007. As the worm turns: The earthworm gut as a transient habitat for soil 598
microbial biomes. Annu. Rev. Microbiol. 61, 169–189.
599
Fahey, T.J., Yavitt, J.B., Sherman, R.E., Maerz, J.C., Groffman, P.M., Fisk, M.C., Bohlen, P.J., 2013.
600
Earthworm effects on the incorporation of litter C and N into soil organic matter in a sugar 601
maple forest. Ecol. Appl. 23, 1185–1201.
602
Fontaine, S., Barot, S., Barre, P., Bdioui, N., Mary, B., Rumpel, C., 2007. Stability of organic carbon in 603
deep soil layers controlled by fresh carbon supply. Nature 450, 277–280.
604
Fontaine, S., Henault, C., Aamor, A., Bdioui, N., Bloor, J.M.G., Maire, V., Mary, B., Revaillot, S., Maron, 605
P.A., 2011. Fungi mediate long term sequestration of carbon and nitrogen in soil through 606
their priming effect. Soil Biol. Biochem. 43, 86–96.
607
Fonte, S., Quintero, D.C., Velásquez, E., Lavelle, P., 2012. Interactive effects of plants and earthworms 608
on the physical stabilization of soil organic matter in aggregates. Plant Soil 359, 205–214.
609
Fonte, S.J., Kong, A.Y.Y., van Kessel, C., Hendrix, P.F., Six, J., 2007. Influence of earthworm activity on 610
aggregate-associated carbon and nitrogen dynamics differs with agroecosystem 611
management. Soil Biol. Biochem. 39, 1014–1022.
612
Franzluebbers, A.J., Haney, R.L., Honeycutt, C.W., Schomberg, H.H., Hons, F.M., 2000. Flush of carbon 613
dioxide following rewetting of dried soil relates to active organic pools. Soil Sci. Soc. Am. J.
614
64, 613–623.
615
Frouz, J., Krištůfek, V., Livečková, M., van Loo, D., Jacobs, P., Van Hoorebeke, L., 2011. Microbial 616
properties of soil aggregates created by earthworms and other factors: spherical and 617
prismatic soil aggregates from unreclaimed post-mining sites. Folia Microbiol. (Praha) 56, 36–
618
43.
619
Fründ, H.-C., Butt, K., Capowiez, Y., Eisenhauer, N., Emmerling, C., Ernst, G., Potthoff, M., Schädler, 620
M., Schrader, S., 2010. Using earthworms as model organisms in the laboratory:
621
Recommendations for experimental implementations. Pedobiologia 53, 119–125.
622
Gaillard, V., Chenu, C., Recous, S., Richard, G., 1999. Carbon, nitrogen and microbial gradients 623
induced by plant residues decomposing in soil. Eur. J. Soil Sci. 50, 567–578.
624
27 Gale, W.J., Cambardella, C.A., 2000. Carbon Dynamics of Surface Residue– and Root-derived Organic 625
Matter under Simulated No-till1Reference to trade names and companies is made for 626
information purposes only and does not imply endorsement by the USDA. Soil Sci Soc Am J 627
64, 190–195.
628
Gleixner, G., Czimczik, C.J., Kramer, C., Lühker, B., Schmidt, M.W.I., 2001. 1.15 - Plant Compounds and 629
Their Turnover and Stabilization as Soil Organic Matter, in: Schulze, E.-D., Heimann, M., 630
Harrison, S., Holland, E., Lloyd, J., Prentice, I.C., Schimel, D. (Eds.), Global Biogeochemical 631
Cycles in the Climate System. Academic Press, San Diego, pp. 201–215.
632
Harris, D., Horwáth, W.R., van Kessel, C., 2001. Acid fumigation of soils to remove carbonates prior to 633
total organic carbon or Carbon-13 isotopic analysis. Soil Sci Soc Am J 65, 1853–1856.
634
Hong, H.N., Rumpel, C., Henry des Tureaux, T., Bardoux, G., Billou, D., Tran Duc, T., Jouquet, P., 2011.
635
How do earthworms influence organic matter quantity and quality in tropical soils? Soil Biol.
636
Biochem. 43, 223–230.
637
Hopkins, D.W., Marinari, S., Tilston, E.L., Halpin, C., 2005. Lumbricus terrestris counteract the effects 638
of modified lignin biosynthesis on the decomposition of tobacco plant residues. Soil Biol.
639
Biochem. 37, 1141–1144.
640
Jégou, D., Cluzeau, D., Balesdent, J., Tréhen, P., 1998. Effects of four ecological categories of 641
earthworms on carbon transfer in soil. Appl. Soil Ecol. 9, 249–255.
642
Jégou, D., Cluzeau, D., Hallaire, V., Balesdent, J., Tréhen, P., 2000. Burrowing activity of the 643
earthworms Lumbricus terrestris and Aporrectodea giardi and consequences on C transfers in 644
soil. Eur. J. Soil Biol. 36, 27–34.
645
Jiménez, J.J., Moreno, G.A., Decaëns, T., Lavelle, P., Fisher, J.M., Thomas, J.R., 1998. Earthworm 646
communities in native savannas and man-made pastures of the Eastern Plains of Colombia.
647
Biol. Fertil. Soils 28, 101–110.
648
Jobbágy, E.G., Jackson, R.B., 2000. The vertical distribution of soil carbon and its relation to climate 649
and vegetation. Ecol. Appl. 10, 423–436.
650
Kammer, A., Hagedorn, F., 2011. Mineralisation, leaching and stabilisation of 13$C-labelled leaf and 651
twig litter in a beech forest soil. Biogeosciences 8, 2195–2208.
652
Klumpp, K., Soussana, J.F., Falcimagne, R., 2007. Long-term steady state 13C labelling to investigate 653
soil carbon turnover in grasslands. Biogeosciences 4, 385–394.
654
Kögel-Knabner, I., 2002. The macromolecular organic composition of plant and microbial residues as 655
inputs to soil organic matter. Soil Biol. Biochem. 34, 139–162.
656
Kuzyakov, Y., 2010. Priming effects: Interactions between living and dead organic matter. Soil Biol.
657
Biochem. 42, 1363–1371.
658
Kuzyakov, Y., Blagodatskaya, E., 2015. Microbial hotspots and hot moments in soil: Concept & review.
659
Soil Biol. Biochem. 83, 184–199.
660
Lavelle, P., Martin, A., 1992. Small-scale and large-scale effects of endogeic earthworms on soil 661
organic matter dynamics in soils of the humid tropics. Soil Biol. Biochem. 24, 1491–1498.
662
Lavelle, P., Blanchart, E., Martin, A., Martin, S., Spain, A., 1993. A Hierarchical model for 663
decomposition in terrestrial ecosystems: application to soils of the humid tropics. Biotropica 664
25, 130–150.
665
Lavelle, P., Lattaud, C., Trigo, D., Barois, I., 1995. Mutualism and biodiversity in soils, in: The 666
Significance and Regulation of Soil Biodiversity. Springer, pp. 23–33.
667
Lee, K.E., 1985. Earthworms: their ecology and relationships with soils and land use. Academic Press, 668
Sydney ; Orlando.
669
Lu, Y., Watanabe, A., Kimura, M., 2003. Carbon dynamics of rhizodeposits, root- and shoot-residues 670
in a rice soil. Soil Biol. Biochem. 35, 1223–1230.
671
Machinet, G.E., Bertrand, I., Barrière, Y., Chabbert, B., Recous, S., 2011. Impact of plant cell wall 672
network on biodegradation in soil: role of lignin composition and phenolic acids in roots from 673
16 maize genotypes. Soil Biol. Biochem. 43, 1544–1552.
674
28 Mambelli, S., Bird, J.A., Gleixner, G., Dawson, T.E., Torn, M.S., 2011. Relative contribution of foliar 675
and fine root pine litter to the molecular composition of soil organic matter after in situ 676
degradation. Org. Geochem. 42, 1099–1108.
677
Mariani, L., Jiménez, J.J., Asakawa, N., Thomas, R.J., Decaëns, T., 2007. What happens to earthworm 678
casts in the soil? A field study of carbon and nitrogen dynamics in Neotropical savannahs. Soil 679
Biol. Biochem. 39, 757–767.
680
Martin, A., 1991. Short- and long-term effects of the endogeic earthworm Millsonia anomala 681
(Omodeo) (Megascolecidæ, Oligochæta) of tropical savannas, on soil organic matter. Biol.
682
Fertil. Soils 11, 234–238.
683
Martin, A., Cortez, J., Barois, I., Lavelle, P., 1987. Les mucus intestinaux de ver de terre moteur de 684
leurs interactions avec la microflore. Rev Ecol Biol Sol 24, 549–558.
685
Nguyen Tu, T.T., Egasse, C., Zeller, B., Bardoux, G., Biron, P., Ponge, J.-F., David, B., Derenne, S., 2011.
686
Early degradation of plant alkanes in soils: A litterbag experiment using 13C-labelled leaves.
687
Soil Biol. Biochem. 43, 2222–2228.
688
Oades, J.M., 1988. The retention of organic matter in soils. Biogeochemistry 5, 35–70.
689
Parle, J.N., 1963. A Microbiological Study of Earthworm Casts. J. Gen. Microbiol. 31, 13–22.
690
Puget, P., Drinkwater, L.E., 2001. Short-Term Dynamics of Root- and Shoot-Derived Carbon from a 691
Leguminous Green Manure. Soil Sci Soc Am J 65, 771–779.
692
Rasse, D., Rumpel, C., Dignac, M.-F., 2005. Is soil carbon mostly root carbon? Mechanisms for a 693
specific stabilisation. Plant Soil 269, 341–356.
694
Rubino, M., Dungait, J.A.J., Evershed, R.P., Bertolini, T., De Angelis, P., D’Onofrio, A., Lagomarsino, A., 695
Lubritto, C., Merola, A., Terrasi, F., Cotrufo, M.F., 2010. Carbon input belowground is the 696
major C flux contributing to leaf litter mass loss: Evidences from a 13C labelled-leaf litter 697
experiment. Soil Biol. Biochem. 42, 1009–1016.
698
Scheu, S., Schauermann, J., 1994. Decomposition of roots and twigs: Effects of wood type (beech and 699
ash), diameter, site of exposure and macrofauna exclusion. Plant Soil 163, 13–24.
700
Shi, A., Penfold, C., Marschner, P., 2013. Decomposition of roots and shoots of perennial grasses and 701
annual barley—separately or in two residue mixes. Biol. Fertil. Soils 49, 673–680.
702
Shipitalo, M.J., Protz, R., 1989. Chemistry and micromorphology of aggregation in earthworm casts.
703
Geoderma 45, 357–374.
704
Six, J., Bossuyt, H., Degryze, S., Denef, K., 2004. A history of research on the link between 705
(micro)aggregates, soil biota, and soil organic matter dynamics. Adv. Soil Struct. Res. 79, 7–
706
31.
707
Steffens, C., Helfrich, M., Joergensen, R.G., Eissfeller, V., Flessa, H., 2015. Translocation of 13C- 708
labeled leaf or root litter carbon of beech (Fagus sylvatica L.) and ash (Fraxinus excelsior L.) 709
during decomposition – A laboratory incubation experiment. Soil Biol. Biochem. 83, 125–137.
710
Stockmann, U., Adams, M.A., Crawford, J.W., Field, D.J., Henakaarchchi, N., Jenkins, M., Minasny, B., 711
McBratney, A.B., Courcelles, V. de R. de, Singh, K., Wheeler, I., Abbott, L., Angers, D.A., 712
Baldock, J., Bird, M., Brookes, P.C., Chenu, C., Jastrow, J.D., Lal, R., Lehmann, J., O’Donnell, 713
A.G., Parton, W.J., Whitehead, D., Zimmermann, M., 2013. The knowns, known unknowns 714
and unknowns of sequestration of soil organic carbon. Agric. Ecosyst. Environ. 164, 80–99.
715
Stromberger, M.E., Keith, A.M., Schmidt, O., 2012. Distinct microbial and faunal communities and 716
translocated carbon in Lumbricus terrestris drilospheres. Soil Biol. Biochem. 46, 155–162.
717
Swift, M.J., Heal, O.W., Anderson, J.M., 1979. Decomposition in terrestrial ecosystems. Univ of 718
California Press.
719
Tahir, M.M., Recous, S., Aita, C., Schmatz, R., Pilecco, G.E., Giacomini, S.J., 2016. In situ roots 720
decompose faster than shoots left on the soil surface under subtropical no-till conditions.
721
Biol. Fertil. Soils 52, 853–865.
722
Taylor, B.R., Parkinson, D., Parsons, W.F., 1989. Nitrogen and lignin content as predictors of litter 723
decay rates: a microcosm test. Ecology 70, 97–104.
724
Thompson, R.B., 1996. Pulse-labelling a cover crop with 13C to follow its decomposition in soil under 725
field conditions. Plant Soil 180, 49–55.
726
